Intel has fixed a pair of flaws in its chips ahead of a planned demonstration of remote attacks on them by security researcher Kris Kaspersky.

Kaspersky – no relation to the Russian security firm – was due to demonstrate the findings of his CPU malware research at the Hack in the Box conference Malaysia in October, in a presentation entitled Remote Code Execution Through Intel CPU Bugs.

The abstract for the presentation noted that the Intel Core 2 architecture has 128 confirmed bugs, with the Itanium packing over 230, and that some of these were vulnerable to local and remote attacks under any OS.

Kaspersky had promised to demonstrate how “it’s possible to make an attack via JavaScript code or just TCP/IP packets storms against Intel based machines. Some of the bugs that will be shown are exploitable via common instruction sequences and by knowing the mechanics behind certain JIT Java-compilers, attackers can force the compiler to do what they want (for example short nested loops lead to system crashes on many CPUs).”

He also promised to show how his experience in data recovery had revealed “how CPU bugs have actually contributed in damaging our hard drives without our knowledge”.

All of which would have added up to a major headache for Intel, and by extension the millions of consumers and customers that depend on its kit - except perhaps for the Itanic.

Sadly or luckily, depending which way you look at it, Kaspersky told ComputerWorld late last week that Intel told him it had now patched two bugs which allowed remote attacks. He added that a dozen other non-remote executable bugs would remain unpatched.

He added that he had been asked not to make proof of concept code available or demonstrate some of his hacks.

All of which makes his presentation somewhat less of a hot ticket than when the agenda was originally posted.

It also casts Intel's denials of a warning by Theo de Raadt that flaws in the Core 2 architecture could be exploited in a new light. ®

More comments…